Odor information is generally assumed to be encoded by olfactory receptor neurons (ORNs) in which odorants evoke a tonic discharge. Exciting new evidence extends hints in the earlier literature that some ORNs are not tonically active, but rather are inherently rhythmically active or 'bursting' ORNs (bORNs) in which the rhythmic bursting is entrained by odorants. bORNs open up an entirely new possibility for encoding odor information, in particular information relative to the spatiotemporal characteristics of the odor signal. Experiments are proposed using an animal model in which both tonically active and bORNs are well characterized physiologically to compare the potential of both types of ORNs to encode odor information. This will be done by first using computational modeling to test the hypothesis that synchronization of large ensembles of bORNs selectively favors (1) the detection and amplification of weak odor signals and (2) the detection and coding of the temporal characteristics of the odor signal itself compared to tonically active ORNs. Combined optical and electrophysiological recording from small ensembles of actual ORNs will then be used to experimentally test the predictions of the computational model. This project has the capacity to identify a heretofore unappreciated way in which olfactory information is encoded.